Artificial Intelligence (AI) is considered as one of the new technologies with the highest potential to transform the financial sector in the near future. AI is increasingly used by fintechs and more established financial firms to improve the efficiency of their operations and of customer-facing activities and also to provide customers with more tailored products and improved service. New generations of AI such as generative AI and greater analytical capacity supported by cloud services and high performance computing should help to accelerate the uptake of AI going forward. Efforts however remain to be made to improve data quality and standardisation to allow AI systems to be trained and function effectively. In addition, the ethical and safety challenges in terms of bias, explainability and rights protection potentially raised by AI need tackling to ensure sufficient customer trust.

A cross-sectoral AI Act was proposed by the European Commission in April 2021, aiming to ensure that AI systems used in the EU are safe and respect fundamental rights and EU values. It follows a risk-based approach imposing specific requirements (in terms of human oversight, documentation, risk assessment…) for AI systems considered as ‘high risk’, while transparency requirements will be imposed on other systems.

At present, AI-based credit scoring and creditworthiness assessment systems and risk assessments and pricing in life and health insurance are considered as ‘high-risk’ in the financial sector. Specific transparency requirements have also been added in the AI Act regarding Generative AI systems: disclosure that content was generated by AI, model for preventing the generation of illegal content and publication of summaries of copyrighted data used for training.

The trilogue negotiations on the AI Act proposal are due to be finalised by the end of 2023, following the general approach agreed at Council level in December 2022 and the adoption of the European Parliament’s report in June 2023.

Policy work on AI is also underway in other major jurisdictions such as the UK and US and also at the international level led in particular by the OECD, which published the first set of international principles on AI in May 2019, aiming to promote innovative and trustworthy AI.